Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for changing a pattern in data according to at least one data model in a communication system comprising hardware, said data comprising collected data of activities of user terminals communicating with the communication system using a network interface and said data of activities being grouped in a data base according to user terminal, the method being performed in a pattern processing node of the communication system and comprising the steps of: obtaining an existing pattern of historical data according to at least one existing data model, said existing pattern relating to said user terminal associated with the communication system, obtaining a further pattern of newer data according to a further data model, said further pattern also relating to said user terminal, wherein the historical data is generated before the existing pattern is identified and the newer data is generated after the historical data, comparing the existing pattern with the further pattern, determining if the existing data model can be mapped on the further data model based on the comparison, determining a Euclidean distance between a vector of the existing pattern and a vector of the further pattern; when the Euclidean distance between the vector of the existing pattern and the vector of the further pattern is above a distance threshold, updating the existing pattern with the further pattern in relation to the historical data based on determining that the existing data model can be mapped on the further data model, wherein the updated existing pattern is stored in the database; and when the Euclidean distance is not above the distance threshold, the existing pattern is not updated.
A method for updating data patterns related to user activity in a communication system. It involves collecting data about user terminals communicating through a network, grouping this data in a database by user terminal, and performing the update in a pattern processing node. The method obtains an existing pattern of historical data for a user, based on an existing data model, and a newer pattern of more recent data for the same user, based on a further data model. It compares these patterns, determines if the existing data model can be mapped to the further data model. A Euclidean distance is calculated between the existing and further patterns. If the distance exceeds a threshold and the models map, the existing pattern is updated with the newer pattern and stored in the database; otherwise, the existing pattern remains unchanged.
2. The method according to claim 1 , wherein each pattern comprises a number of attributes, the step of comparing patterns comprises comparing attributes of the patterns with each other, where each attribute corresponds to data of a data model and the step of updating comprises adding attributes of the further pattern to the existing pattern.
The method for updating data patterns, as described, where each pattern consists of attributes. Comparing the patterns involves comparing the attributes of each pattern. Each attribute corresponds to a piece of data from a data model. Updating the pattern involves adding attributes from the newer pattern to the existing pattern, effectively enriching the existing pattern with new information derived from the newer data. This allows the system to adapt the data model over time to better reflect evolving user behaviors.
3. The method according to claim 2 , wherein each pattern is provided as a vector and the step of comparing comprises determining the Euclidean distance between the vectors and the step of determining if the existing data model can be mapped on the further data model comprises comparing the Euclidean distance with the distance threshold.
The method for updating data patterns, as described where patterns are represented as vectors. Comparing patterns involves calculating the Euclidean distance between these vectors. Determining if the existing data model can be mapped to the further data model, is done by comparing the calculated Euclidean distance with a predefined distance threshold. This threshold determines the degree of similarity required for the existing pattern to be updated with the newer pattern, allowing a quantitative assessment of model compatibility.
4. The method according to claim 1 , wherein the step of comparing is made while considering a set of policies.
The method for updating data patterns, as described where the pattern comparison process takes into account a set of predefined policies. This means that the comparison of the existing and newer patterns isn't solely based on the data models themselves, but also on rules or guidelines that dictate how certain attributes or relationships should be handled during the comparison. These policies can influence whether the existing pattern is ultimately updated.
5. The method according to claim 4 , wherein the set of policies comprises an absolute policy regarding at least one attribute.
The method for updating data patterns considering policies, as described, includes an "absolute policy." This policy type sets fixed, non-negotiable rules for at least one attribute within the data patterns. These absolute policies will determine if specific data attributes in the patterns, cannot be updated, overwritten or changed in certain situations.
6. The method according to claim 4 , wherein the set of policies comprises an aggregate policy regarding at least one attribute.
The method for updating data patterns considering policies, as described, includes an "aggregate policy." This policy type applies rules to aggregated data or summaries of data attributes. Aggregate policies will determine if the update should be allowed if certain aggregated values within the patterns, meet some requirements.
7. The method according to claim 1 , wherein the step of comparing the existing pattern with the further pattern comprises comparing the occurrence of the further pattern in relation to the occurrence of the existing pattern in the newer data according to the further data model and the step of updating comprises updating the existing pattern with the further pattern with the same relationship as in the newer data.
The method for updating data patterns, as described where the comparison of the existing pattern and the newer pattern focuses on how often the newer pattern occurs relative to the existing pattern within the newer data. Updating the pattern involves updating the existing pattern to reflect the same relative occurrence of the newer pattern as observed in the newer data. This aims to preserve the statistical relationships between patterns during the update process.
8. The method according to claim 1 , further comprising the step of identifying the further pattern based on at least some newer data.
The method for updating data patterns, as described, further includes the step of identifying the newer pattern from the newer data. This means that the method first analyzes the newer data to discover and define the newer pattern before performing any comparisons or updates. This pattern identification step ensures that the method can adapt to evolving data characteristics and discover new patterns dynamically.
9. The method according to claim 1 , wherein the existing pattern has a label and further comprising the step of changing the label based on the updated pattern.
The method for updating data patterns, as described where the existing pattern has a label associated with it. The method also includes changing the label based on the updated pattern. This updated label, reflects the updated pattern.
10. A communication system comprising a pattern analysing device in a pattern processing node and a database having data comprising collected data of activities of user terminals communicating with the communication system using a network interface and said data of activities being grouped according to user terminal, the pattern analysing device comprising: a pattern updating unit equipped with a pattern collecting element configured to: obtain an existing pattern of historical data according to at least one existing data model, said existing pattern relating to a user terminal associated with the communication system, and obtain a further pattern of newer data according to a further data model, said further pattern also relating to said user terminal, wherein the historical data is generated before the existing pattern is identified and the newer data is generated after the historical data, compare the existing pattern with the further pattern, and a pattern updating element configured to determine if the existing data model can be mapped on the further data model based on the comparison, and determine a Euclidean distance between a vector of the existing pattern and a vector of the further pattern; when the Euclidean distance between the vector of the existing pattern and the vector of the further pattern is above a distance threshold, update the existing pattern with the further pattern in relation to the historical data based on determining that the existing data model can be mapped on the further data model, wherein the updated existing pattern is stored in the database, and when the Euclidean distance is not above the distance threshold, the existing pattern is not updated.
A communication system for updating data patterns includes a database storing user activity data grouped by user terminal. The system also has a pattern analyzing device in a pattern processing node. This device has a pattern updating unit with a pattern collecting element that obtains an existing pattern of historical data for a user, based on an existing data model, and a newer pattern of more recent data for the same user, based on a further data model. The pattern updating unit compares the patterns, determines if the existing data model can be mapped to the further data model. The pattern updating unit also determines a Euclidean distance between the existing and further patterns. If the distance exceeds a threshold and the models map, the existing pattern is updated with the newer pattern and stored in the database; otherwise, the existing pattern remains unchanged.
11. A communication system according to claim 10 , wherein the pattern analysing device further comprises a pattern analysing unit configured to identify at least the further pattern based on at least some of the newer data.
The communication system for updating data patterns, as described, includes a pattern analyzing unit. This unit identifies the newer pattern from the newer data. This identifies the newer pattern before comparing/updating.
12. The method of claim 1 , wherein the existing pattern is maintained in a pattern database, the method further comprising: responsive to updating the existing pattern, recursively updating the pattern database.
The method for updating data patterns, as described where the existing pattern is stored in a pattern database. After updating the existing pattern, the pattern database is recursively updated. This ensures that all data and patterns within the data base are synchronized and consistent.
14. The communication system of claim 10 , wherein the existing pattern is maintained in a pattern database, the a pattern updating unit equipped with a pattern collecting element further configured to: responsive to updating the existing pattern, recursively update the pattern database. updating the existing pattern.
The communication system for updating data patterns, as described, where the existing pattern is stored in a pattern database. The pattern updating unit recursively updates the pattern database, after updating the existing pattern. This means that after an existing pattern is updated, the pattern database is also updated to reflect the changes.
Unknown
October 31, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.